From 9b4e9cc8a0311e5243d69b73ed073e7ea441982e Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 27 三月 2024 16:05:29 +0800
Subject: [PATCH] train update
---
funasr/utils/load_utils.py | 19 ++++++++++++-------
1 files changed, 12 insertions(+), 7 deletions(-)
diff --git a/funasr/utils/load_utils.py b/funasr/utils/load_utils.py
index 5dc9bde..8ff7115 100644
--- a/funasr/utils/load_utils.py
+++ b/funasr/utils/load_utils.py
@@ -19,11 +19,9 @@
def is_ffmpeg_installed():
try:
- # 灏濊瘯杩愯ffmpeg鍛戒护骞惰幏鍙栧叾鐗堟湰淇℃伅
output = subprocess.check_output(['ffmpeg', '-version'], stderr=subprocess.STDOUT)
return 'ffmpeg version' in output.decode('utf-8')
except (subprocess.CalledProcessError, FileNotFoundError):
- # 鑻ヨ繍琛宖fmpeg鍛戒护澶辫触锛屽垯璁や负ffmpeg鏈畨瑁�
return False
use_ffmpeg=False
@@ -31,7 +29,7 @@
use_ffmpeg = True
else:
print("Notice: ffmpeg is not installed. torchaudio is used to load audio\n"
- "If you want use ffmpeg backend to load audio, please install it by:"
+ "If you want to use ffmpeg backend to load audio, please install it by:"
"\n\tsudo apt install ffmpeg # ubuntu"
"\n\t# brew install ffmpeg # mac")
@@ -53,13 +51,20 @@
if isinstance(data_or_path_or_list, str) and os.path.exists(data_or_path_or_list): # local file
if data_type is None or data_type == "sound":
- if use_ffmpeg:
- data_or_path_or_list = _load_audio_ffmpeg(data_or_path_or_list, sr=fs)
- data_or_path_or_list = torch.from_numpy(data_or_path_or_list).squeeze() # [n_samples,]
- else:
+ # if use_ffmpeg:
+ # data_or_path_or_list = _load_audio_ffmpeg(data_or_path_or_list, sr=fs)
+ # data_or_path_or_list = torch.from_numpy(data_or_path_or_list).squeeze() # [n_samples,]
+ # else:
+ # data_or_path_or_list, audio_fs = torchaudio.load(data_or_path_or_list)
+ # if kwargs.get("reduce_channels", True):
+ # data_or_path_or_list = data_or_path_or_list.mean(0)
+ try:
data_or_path_or_list, audio_fs = torchaudio.load(data_or_path_or_list)
if kwargs.get("reduce_channels", True):
data_or_path_or_list = data_or_path_or_list.mean(0)
+ except:
+ data_or_path_or_list = _load_audio_ffmpeg(data_or_path_or_list, sr=fs)
+ data_or_path_or_list = torch.from_numpy(data_or_path_or_list).squeeze() # [n_samples,]
elif data_type == "text" and tokenizer is not None:
data_or_path_or_list = tokenizer.encode(data_or_path_or_list)
elif data_type == "image": # undo
--
Gitblit v1.9.1